This Sequence Analysis Consortium (SAC) brings together experts in Alzheimer's disease (AD), human genetics, biostatistics and genome sciences to identify risk and protective single nucleotide and copy number variants (SNVs and CNVs) influencing AD and AD-related endophenotypes. This SAC will harmonize and integrate genomic and phenotype data (generated by Core C) from the AD Sequencing Project (ADSP) and other data resources and use the most informative analytic methods (as determined by Core B) to identify novel risk and protective alleles. This SAC is partially based on the ADSP case-control study design with whole exome sequencing in 6,888 cases who developed AD despite an 'a priori low risk' and are likely enriched for novel risk variants, and 14,400 cognitivly normal older controls, a subset of whom are enriched for protective variants. To allow novel discovery in minority populations that carry a disproportionate burden of AD, we will analyze exome sequence data and well-characterized endophenotypes, including verbal memory and hippocampal volume (harmonized by Core D). Replication will be afforded using targeted sequence and genotype data in 50,000 cases and controls and 40,000 individuals with endophenotype data. We will seek to identify SNVs (Project 1) and CNVs (Project 2) contributing to AD susceptibility and protection. Integrated and annotated (using ENCODE and the Epigenomic Roadmap Projects) SNVs and CNVs data from the ADSP families with whole genome sequencing, as well as other familial AD resources will be used for genetic linkage and IBD analyses (Project 3) to identify AD susceptibility or protective alleles in highly-ascertained, multiplex families. The case-control design (Projects 1 and 2) and the family design (Project 3) complement each other to maximize discovery of AD variants. In order to promote AD translational research and discovery, we will collaborate with other researchers to undertake further replication and functional studies, create public resources with integrated genomic and phenotypic data, and make available computational algorithms through the public domain. Identification of variants in genes leading to risk and protection from AD may ultimately lead to novel treatments and prevention strategies.